# Dowel Maker Hewn Tool Library · https://hewn.build/tools/dowel-maker Corrected: An independent reviewer read this entry and found corrections. They have been applied to the text, and the reviewer re-checked the correction that mattered most for safety and found it done; the others have not been re-checked yet. The list at the end is the record of what changed. Cite as: "Dowel Maker." Hewn Tool Library, hewn.build/tools/dowel-maker. Reviewed 2026-07-29. --- ## identity **Name**: Dowel maker (family). Form one: **dowel plate** (also dowel-cutting plate, peg plate; Veritas sells a bladed hybrid as the "Dowel Former"). Form two: **rotary dowel maker** (Veritas Dowel Maker; historically the Stanley No. 77 "dowel and rod turning machine"). Adjacent lineage: **rounding plane**, with the richest alternate-name list in this corpus so far — per Lee Valley's own tool-history column: *witchet, trapping plane, stail engine, turning head, forkstaff rounder, trenail rounder (also moot, nogg, or nugg), thole plane* [D7]. **Family / category**: Dowel- and rod-making tools — sizing/rounding tools that convert straight-grained stock into cylinders of exact diameter; siblings of the drawplate in metalworking by analogy only (see *lineage-misuse* — the analogy is exactly where the tool gets misused). **One line**: A tool that makes a round, exactly-sized dowel or peg out of wood *you* chose — by shearing a riven blank through a hardened plate, or spinning a square blank past a pair of blades — because the dowel you can buy may be the wrong size, species or grain for work that matters. --- ## ✦ how NOT to use it ### Not-for: jobs it should never do - **Long or show-surface dowels through a plate** are usually easier and more repeatable with a rotary maker, lathe or commercial stock; this is a practicality preference, not an absolute prohibition. The plate has no lateral guidance — the dowel follows the strikes, and past ~6" it came out bowed and faceted in one user's report (Trebuna [D5]). Allan Brown's honest frame from the founding thread: perfect roundness "may be achievable, I hardly think it worth the effort... all perfectly serviceable" — for *pegs*; not for exposed precision dowels [D4]. - **Softwood in the plate, and softwood structural pegs, are conditions to check, not blanket prohibitions.** Cedar crumbled in the founding thread (one user's experience [D4]), and a practitioner there says a softwood peg may not take enough drawbore offset ("you won't be able to offset enough," Harlan Barnhart [D4]). Whether a softwood peg is suitable depends on peg diameter, species and density, member material, joint geometry, moisture, load and desired failure mode. For furniture, timber framing, restoration or any consequential loaded joint, use a tradition-, test-, plan- or engineer-supported peg specification, and verify species, grain, moisture, diameter, hole layout, edge distance and drawbore offset together [Timber Framers Guild curriculum and Forest Products Laboratory Wood Handbook ch. 8, per the review, relayed]. - **Tapered work — chair tenons, handle tapers.** The straight-cylinder tools do not taper. This rotary dowel maker produces cylinders, not intentional tapers. Choose a straight, tapered or wedged tenon system according to the chair design, mating reamer or cutter, stock, load and repair tradition: straight cylindrical tenons are used successfully in chair and furniture work when properly sized and fitted, and the claim that a straight tenon will work loose comes from one enthusiast blog [D6], not from joint doctrine. Use a dedicated tenon tool, rounder or lathe when shoulders, tapers or end-only cutting are required. - **One-off pegs through the rotary maker.** Economics, not physics: setup and price make it a production tool; the split-and-whittle method or the plate serves single pegs faster [D8][D11]. ### Dangerous: misuse that injures the user *Rebuilt in the fix pass. The earlier draft said no injury account was located and inferred hazards from mechanism; that is a research result, not a risk assessment. The controls below come from the official Veritas manual as described in the reviewer's 2026-07-29 review (the manual could not be fetched in this pass, so each manual line is marked `[manual, relayed]`), the CCOHS struck-tools page and Lie-Nielsen's page (both read in this pass). Follow the manufacturer's manual for the exact tool.* **Veritas rotary dowel maker** (a powered, rotating workpiece with exposed stationary blades and several feet of stock): - Secure the body to the bench, in a suitable vise, or with the approved dog-hole assembly [manual, relayed; the mounting options also appear on the Lee Valley page, D1]. - Wear eye protection [manual, relayed]. - Treat both blades as extremely sharp. Never put a finger into either cut-out, including to clear chips [manual, relayed]. - Stop and isolate the drill before clearing chips, adjusting, changing inserts or bushings, or touching blade assemblies [the reviewer's recommended control; manual, relayed]. - Fit the correct insert, guide bushing and socket, and use the manufacturer-specified blank size for the output (see *correct use*) [manual, relayed]. - Align the drill, blank and maker bore coaxially before and during the cut [manual, relayed]. - Use clear, straight stock; knots and poor grain can break on infeed or outfeed [manual, relayed]. - Support the output at the manual's intervals: 1/4" dowel every 8"; over 1/4" through 5/8" every 12"; over 5/8" every 15". Unsupported dowel can whip, affect the finish and break; longer square stock can twist and fail under drive torque; and there is a finite achievable length even with support [manual, relayed]. - Keep hands, clothing, hair, cords, tools and bystanders outside the rotating-stock and whipping radius, and keep people clear of the output path [manual, relayed]. - Use high drill speed with a slow, constant feed. Do not force a seized blank or start-stop it to free a jam [manual, relayed]. If a long blank jams in the cutterhead, the torque goes to the drill and to the free end of the whipping blank (mechanism). - Stop if seizing, uncontrolled vibration, whip, twist, cracking or loss of axial control develops. **Dowel plate** (hardened A2 steel at about 60 RC, struck with a hammer while a blank is driven through it): - Mount or clamp the plate over a clear relief hole, never over a hand, leg or unsupported surface. Lie-Nielsen describes screwing it to the bench through the mounting holes or clamping it over a dog hole [D2, read this pass]. Held in one hand or braced over the knees it moves under the strike; practitioner answers are a bored block or an angled vise fixture at belt height (Hauser, Nickerson [D4]). - Wear eye protection and protect bystanders from chips or fragments. CCOHS recommends safety glasses, goggles or a face shield, and screens for nearby workers [D17, read this pass]. - Prepare the blank near the target diameter and chamfer the lead end so it starts under control [D2]. - Strike the wood blank squarely with an appropriate, sound striking tool, and never intentionally strike the hardened plate. Hardened steel struck by hardened steel is the chip and spall scenario that struck-tool guidance addresses. CCOHS also says to check striking heads for mushrooming or chips and discard cracked or chipped tools; it does not specifically prescribe a wooden or dead-blow mallet for this plate, so a wooden mallet or dead-blow on the blank is this entry's practical judgment, not CCOHS doctrine [D17]. - Keep the supporting hand away from the strike zone and the emerging end. Stop if the blank splits, wedges or rebounds; do not catch or steady it directly below the hole. - Keep the cutting surface sharp, clean and lightly oiled against rust [D2]. ### Damaging: misuse that ruins the tool or the work - **Feeding the rotary maker too fast, or with undersized/unsheared stock — the documented disaster.** scrollsawguy's "ragged tornout lookin stick" [D9] is the family's Sawmill-Creek-Disaster equivalent, and its named diagnosis is a checklist: feed speed, blade sharpness, stock the right size for the output (the manual's table, see *correct use*; Slik Geek's checklist said 1/8" oversize), and octagonal prep — "trim the stock down on 8 sides, instead of 4" to unload the roughing blade, especially in hard or soft-tearing species (Uncle Cracker) [D9]. - **Asking one plate hole to remove heavy stock.** It can crush fibers and jam blanks; Lie-Nielsen's instruction is to prepare the blank near the finished size and use the appropriate hole [D2]. "Start by tapping it through the larger sizes and work your way down... a little off at a time" (Charles Murray [D4]) is a practitioner technique; if used, keep reductions modest. - **No chamfer on the lead end.** Lie-Nielsen's own instruction — "chamfer the leading end of the blank with a knife to make it easy to start the cut" [D2]; an unchamfered blank mushrooms at the entry and wedges. - **Setting the finishing blade without a test section.** Springback differs by species; Lee Valley's own procedure is cut-short-test-then-adjust [D1]. Setting once and switching species quietly changes every diameter after. - **Hammering askew.** "Hit a nail on one side and it bends away from the blow, same with making dowels with a plate" (Strasil [D4]) — crooked strikes make crooked pegs and wallow the hole entry. - **Letting the plate rust.** "Keep oiled" — the cutting lip is the top face; rust on it is edge damage [D2]. ### Beginner-trap: mistakes that look fine until they aren't - **★ Dowel stock with hidden grain runout — the trap that can break joints, not just pegs.** A dowel with grain runout can look identical to one with straight grain, and shape alone does not reveal how it was made; a sawn blank can also be selected for straight grain. Under drawbore offset — or years of load — a run-out peg can shear along its own grain lines. Inspect or specify the grain rather than assume every purchased or sawn dowel is compromised. Named practitioners approach this from different directions: Schwarz (riven is stronger, choose dowel stock only "with straight grain" [D10]), Maguire (riven grain "parallel in both planes" [D8]), Mickley (a species-mismatch rule of thumb: "a white oak pin will just blow right through a white pine joint" [D13-thread]) and Tom M King (the historical record of pegs blowing out tenons [D13-thread]). These are practitioner statements, not tested design values. - **Expecting commercial-dowel smoothness from a plate.** The founding thread exists because of this trap: Peet's "not too straight. Or smooth. Or round." The fix was technique *and* expectations — "lowering my expectations" (Nickerson), "perfectly serviceable" (Brown) [D4]. The plate is best suited to short pegs, tenons and sizing; Lie-Nielsen markets exactly sized dowels and says small facets disappear quickly, so finish depends on blank preparation, species, length, strike alignment and sharpness [D2]. - **The inverse trap: trusting the bought dowel without measuring it.** A hardware-store dowel may not match its nominal size, roundness or species. The patent's background states the inventors' view that they "vary wildly... often oval" [D3], and users report measuring variation [D5], but practitioner reports show that undersized or oval dowels occur, not that commercial dowel is generally defective. Measure the dowel and the hole under the relevant shop conditions, and inspect grain for the intended load. - **Too much drawbore offset for the peg's flex.** Offset past what the peg can bend can crack it inside the joint, invisibly: "Rigid, short-grained wood is likely to crack" (Cohen [D13-thread]). Practitioner figures run from 1/16" (Moxon's shilling, via Schwarz) to 3/32" at furniture scale, and 1/32" for stiff pegs (Schwarz [D10]; Koepke [D13-thread]); these are one maker's scale-specific figures, not design values. Verify species, grain, moisture, diameter, hole layout, edge distance and drawbore offset together, and do not transfer a forum offset or species pairing between scales. - **Assuming the rotary maker works out of the box.** The instructions are famously thin ("I'm glad it was not just me who struggled to find useful instructions" [D11]); the test-section blade ritual is not optional. ### Lineage-misuse: treating it like a relative it isn't - **★ Treating the dowel plate like a drawplate.** The metalworker's drawplate *draws* — ductile metal stretches and flows through the die. Wood cannot flow; the dowel plate cuts and shears the blank, with some compression and scraping depending on stock and preparation. Lie-Nielsen calls the top face the cutting surface, and the holes carry a sharpenable lip and 6° relief [D2]; a forum practitioner calls it "a scraping action not cutting" (Murray [D4]). That is why stock should arrive near-size, chamfered and well prepared. Drive a fat square blank through like wire and you can get crushed fibers and a stuck blank. "Drawplate" is a fair analogy only if wood is understood as cut, sheared, compressed and burnished rather than plastically drawn. - **★ Treating any purchased or machine-made dowel as interchangeable with a grain-selected peg.** The round stick is the same shape its riven ancestor was; the grain discipline is what the shape used to mean in loaded work. A dowel dropped into a drawbored joint without checking grain, species, moisture and fit may not suit the joint. Purchased stock is not automatically defective, though, and a sawn or turned blank can be selected for straight, continuous grain; riving makes grain-following selection easier and is often preferred for highly bent pegs [D10][D8]. - **Treating the rotary dowel maker like a pencil sharpener or tenon cutter.** It is not fed round stock to "clean up," and it does not taper; it wants a square blank of the size the manufacturer's table gives for the output, driven as its own axle, cut twice [D1][D3]. Feeding near-size or round stock gives the guides nothing to index and the roughing blade nothing it was set for. - **Treating any dowel maker like a lathe.** No tapers, no swells, no shoulders — the adjustable witchet could taper, slowly and loosely (with documented play [D6]); its precision descendants gave up the taper to gain the accuracy. Asking for it back is asking the tool to un-evolve. --- ## correct use *(brief — the workbench holds deep operation)* - **Plate**: prepare the blank close to the chosen size — riven stock squared then octagoned, or whittled, planed or drawknifed ("get as close as you can to the correct size with a plane, spokeshave or drawknife" — Lie-Nielsen's own instruction; small facets disappear as it cuts); **chamfer the leading end**; pick the finished-size hole you want and drive the blank through it squarely, over a supported relief hole, at belt height [D2][D4]. If a maker elects to reduce progressively through larger holes, that is a practitioner technique; keep reductions modest [D4]. Wax helps (practitioner report) [D4]. Finish depends on blank preparation, species, length, strike alignment and sharpness. The plate is best suited to short pegs, tenons and sizing tasks where manual support and strike control are practical; species and structural suitability are joint-specific (see *how NOT*). - **Rotary maker**: size the square blank from the manufacturer's table for the output size, not from one rule. Per the official Veritas manual the blank is 1/8" oversize for dowels in 1/8" increments and 3/16" oversize for dowels in 1/16" increments (for example a 5/16" dowel starts from 1/2" square stock) [manual, relayed — the review; the manual could not be fetched in this pass]; the Lee Valley page's shorter "rip square stock 1/8" oversize" is a simplification [D1]. Knock the corners to octagon for hard or tearout-prone species (practitioner technique [D9]). Use an electric drill compatible with the supplied 3/8" drive adapter and capable of maintaining the manual's cutting conditions (high drill speed with a slow, constant feed after setup); keep the drill and stock coaxial with the maker bore; follow the drill maker's limits as well as Veritas's instructions [manual, relayed]. A corded 1/2" drill is a practitioner preference, labeled as experience, not a requirement [D9][D11]. **Set the finishing blade by cutting a short test section** and micro-adjusting for the species' springback before committing good stock [D1]. **Support the output at the manual's diameter-specific intervals:** 1/4" dowel every 8"; over 1/4" through 5/8" every 12"; over 5/8" every 15" [manual, relayed]. **Length is manufacturer-rated potential, not a guarantee:** the manual says a finite achievable length exists even with support, governed by species, dryness, knots and the square blank's ability to withstand drive torque [manual, relayed]; maximum practical length depends on diameter, supports, straightness, species, moisture, defects, alignment and torque response. - **Either way, the grain rule outranks the tool**: the blank should be demonstrably straight-grained. Riving makes grain-following selection easier and is often preferred for highly bent drawbore pegs; sawn, turned or purchased dowel must be inspected or specified for sufficiently straight, continuous grain when the application depends on bending or shear capacity [D8][D10]. --- ## ✦ matching *(the workbench's first read)* **does** — makes dowels/pegs/rods of tuned, measured diameter (measured at the relevant moisture condition, since wood moves after machining) from the species the maker chooses, subject to stock limits (straightness, grain continuity, density, openness of grain, knots, moisture, required finish and load all matter; "any species" is manufacturer marketing, and the manual's troubleshooting says open grain, unusual grain orientation, knots and species can give poor finish or breakage [manual, relayed]); makes drawbore pegs from riven stock; sizes slightly-oversize round stock down to true diameter (plate — including Don Dorn's run-a-commercial-dowel-through trick [D4]); makes long rods (rotary; Lee Valley rates dowels to 4 ft "or longer in the larger diameters" [D1], conditional on the support intervals and stock limits in *correct use*); cuts precise round tenons on chair-spindle ends (plate, per Lie-Nielsen's own use list [D2]). **suits** — **draw-bored mortise-and-tenon work** — a principal application: workbenches, doors, frames, timber joinery at furniture scale, where the peg is bent inside the joint and riven or demonstrably straight-grained stock is preferred [D10][D8]; for any consequential loaded joint, use a tradition-, test-, plan- or engineer-supported peg specification; **restoration and repair** — matching the odd diameters and out-of-commerce species of old work, where a hardware-store dowel may be the wrong size, species or grain (the patent background states the inventors' view of this [D3]; Steve Hubbard's padauk/maple motive [D5]); **chairmaking production runs** — 50 straight-grained spindles at a time (Schwarz; Mursell's courses [D11]); **shop-made rods generally** — canes, rails, pulls, pins — where diameter consistency matters (Droege's 3/8" motive [D5]). **kit-fit** — **pairs with**: froe/chisel/drawknife for riving and octagoning the blanks [D4][D8]; a bench hook and wide chisel (Maguire's octagon step [D8]); an electric drill compatible with the supplied 3/8" drive adapter for the rotary form (a corded 1/2" drill is a practitioner preference, not a requirement [D9][D11]); a secured mount (bench, vise or dog-hole assembly) and a way to support the output at the manual's intervals — 8", 12" or 15" by diameter, see *correct use* [manual, relayed]; drawbore pins for test-assembly of the joints the pegs will live in; a [[Spoke-Pointer|spoke pointer]] for starting blanks [D6] — now its own corpus entry (Run 614). **Substitutes for**: the lathe, for plain cylinders (faster per piece, no turning skill); commercial dowel, where the project needs a different species, diameter, fit or controlled grain (purchased dowel is not automatically defective; measure and inspect it); Maguire's split-and-whittle method substitutes for the *whole family* at zero tool cost for peg-scale work [D8]. **Does not replace**: the lathe for tapers, swells, or anything non-cylindrical (that is trapping-plane/lathe territory [D6]); a dedicated tenon tool where shoulders, tapers or end-only cutting are required — this rotary dowel maker produces cylinders, not intentional tapers, and a straight, tapered or wedged tenon system should be chosen from the chair design, mating reamer or cutter, stock, load and repair tradition; commodity dowel for volume at commodity price — nobody feeds a house-trim job through a dowel plate. **not-for** *(project-keyed)* — don't reach for the **plate** for long straight dowels: past roughly 6" they came out bowed in one user's report — "on dowels longer than 6 inches. I can't produce a straight one" (Todd Trebuna [D5]; Lie-Nielsen says any length is possible in principle [D2], so this is a practicality limit, not a prohibition); don't reach for the **rotary maker** for one or two pegs — the setup ritual and price only pay off at quantity; Schwarz's own line for the occasional job is spokeshave, octagon, plane [D11]; don't reach for **either** for tapered chair tenons or handle tapers — the rotary maker makes cylinders, not intentional tapers; use a dedicated tenon tool, rounder or lathe when taper, shoulders or end-only cutting are needed [D6]; softwood in the plate is a condition to check, not a prohibition — cedar crumbled in the founding thread (one user's experience [D4]), and whether a softwood peg suits a structural joint is joint-specific (see *how NOT*) [D4][D13-thread]. **access** — the **plate** is the cheap, beginner-accessible entry (~$100) with a documented expectation trap: first output disappoints everyone (Peet [D4]); with riven-hardwood prep and fair strikes, a beginner makes serviceable pegs day one. The **rotary maker** is beginner-usable mechanically but carries a real setup floor — the test-section blade-setting ritual, notoriously thin instructions ("I'm glad it was not just me who struggled" [D11]), and a $275–495 price that only quantity justifies. The **grain knowledge is the real skill gate for both** — see *beginner-trap*. **revival** — a real revival case, and it is an engineering argument more than nostalgia. Why make dowels when you can buy them: (1) **fit** — the maker may need a tuned, measured diameter for the species and moisture at hand; commercial dowel can vary ("imprecise at best and wildly inaccurate at worst" is Lee Valley's marketing [D1]; the patent background states the inventors' view that commercial dowels vary in size, roundness and species [D3]; users report measuring variation [D5]), though purchased stock is not automatically defective; (2) **species** — commercial dowel comes in a limited range of woods, and restoration, visible joinery and hard-use work may want the project's own species [D3][D5]; (3) **grain** — riving makes grain-following selection easier and is often preferred for highly bent drawbore pegs (Schwarz [D10]; Maguire [D8]), while sawn, turned or purchased dowel must be inspected or specified for sufficiently straight, continuous grain when the application depends on bending or shear capacity; (4) **the peg's role** — in a drawbored joint the peg is a load-carrying and alignment-critical part, and the survival of drawbored antique furniture rests on the whole joint, pegs included [D10]. The museum-to-bench link is not claimed: ancient dowelled artifacts prove early dowel use, while the shaping tool and production method remain unknown [D12]. --- ## anatomy **Dowel plate (Lie-Nielsen standard form)** [D2]: - **Plate body** — 5-1/8" × 1-1/2" × 1/4" or 5/16" (Lie-Nielsen's page states 1/4" in the specification and 5/16" in its instructions and does not resolve the conflict; verify against the physical tool or the maker before settling on one value [D2]), surface-ground A2 tool steel, hardened to 60 RC, cryogenically treated and double-tempered; hardened A2 provides a durable cutting edge; hone the top face when needed and oil it against rust. It is hard enough to matter when struck (see *dangerous*). - **Graduated holes** — seven, 1/8" through 5/8"; the holes are discrete finished target sizes: prepare the blank near the chosen size and drive it through the appropriate hole. Progressive reduction through larger holes is an optional practitioner technique, not inherent anatomy. - **Cutting lip and clearance taper** — each hole is straight for the first .045", then relieved at 6° underneath — Lie-Nielsen calls the top face the cutting surface, and the straight land is the shearing edge and the reason the plate can be resharpened (flatten the top face on stones) without the holes changing size. - **Support** — the plate has no built-in body; it works over a dog hole, a bored block, or angled in a vise (Russ Hauser's fixture [D4]) so the emerging dowel has somewhere to go. **Rotary dowel maker (Veritas 05J45 form)** [D1][D3]: - **Body** — cast, anodized aluminum. - **Front insert guide** — sized square opening that admits and indexes the oversize square blank. - **Roughing blade** — A2 tool steel, aggressive first cut, "rounded trailing corner" per the patent. - **Finishing blade** — A2, fine shearing cut; each blade on its own **micro-adjust assembly**, set by cutting a short test section and dialing to final size — because "fiber springback when cut" differs by species [D1]. - **Rear guide bushings** — steel, sized to the finished dowel plus a 1/16"-undersize spare; the finished dowel's own exit bearing. - **Square socket drive + drill adapter** — the blank itself is the drive shaft: an electric drill spins the square stock through the stationary cutterhead. - **Blade replacement jig** — included; blade position is the tool's whole accuracy. **Rounding plane / [[Witchet (Rounding Plane)|witchet]] (lineage form)** [D6][D7] — now its own dedicated entry, Run 632: - **Tapered round channel** through the body; **angled iron** cutting pencil-sharpener fashion; on adjustable forms ([[Stail Engine]]), **threaded arms** vary the diameter and permit deliberate tapers — and introduce the play that makes them imprecise ("variation in width, due to play in the tool" [D6]). --- ## variations (living) - **Dowel plate, precision commercial** — Lie-Nielsen standard (1/8"–5/8") [D2]; shop-made equivalents (hole drilled in a plane iron or hardened plate) are a documented folk form [D5-thread]. - **Veritas Dowel Maker, basic vs. Master set** — basic (05J4501, $275, pre-drilled 1") vs. Master set (05J4520, $495); range 1/4" to 1" in 1/16" increments, dowels to 4 ft "or longer in the larger diameters" [D1] — manufacturer-rated potential, not a guarantee; see *correct use* for the support intervals and length limit. Made in Canada; patented [D1][D3]. - **Veritas Dowel Former** — the hybrid: hammer-driven like a plate, bladed like the rotary tool [product line, D1-adjacent; not covered in depth this pass]. - **Rounding planes** — fixed rounders; adjustable [[Stail Engine|stail engines]]/witchets; **[[Trapping Plane|trapping planes]]** (lathe-companion form, tapers and swells under tension, now built as its own entry — Run 641) [D6]; Ashem Crafts rotary planes as the living manufactured line [D14]. - **[[Dowel & Tenon Cutter (Hollow Auger)|Dowel & tenon cutters / tapered tenon cutters]]** — end-of-stick siblings, straight and tapered; the tapered form is the chairmaking joint's tool [D6]. --- ## trades - **Furniture-making and joinery** — drawbored M&T, pegged casework; the core living trade [D10][D13-thread]. - **Chairmaking** — spindle/stick production (Veritas in Mursell's Windsor courses [D11]); the rounder tradition's living home (Ashem Crafts [D14]); tenons on spindle ends via plate [D2]. - **Timber framing** — pegged joinery at scale; per Charles Guest, the scale drawboring genuinely suits [D13-thread]. - **Restoration/conservation** — odd diameters, matched species, period-correct pegs [D3][D5]. - **Boatbuilding (historical)** — name-level evidence: *trenail rounder (moot)* and *thole plane* among the witchet's own recorded names [D7]; not independently developed this pass — LOW-MEDIUM. - **Lutherie** — a later patent record (a Wittner GmbH peg-cutter application, 2007) cites the Veritas patent; whether it is a violin-peg descendant is not established [D3]. - **Handle trades (historical)** — brooms, rakes, cues, rods via [[Stail Engine|stail engines]] [D6]; largely industrialized away. --- ## lineage **Ancestors** - **The riven, whittled peg — the true ancestor, and still a living method.** Before any dedicated tool, pegs were split (riven) from straight stock and whittled or shaved round. Richard Maguire documents the unbroken survival of this method: split, square with a chisel, octagon on a bench hook, refine on an inverted hand plane — "A good peg, is a lot more than a roughly ragged out bit of round wood," and riven wood is "almost guaranteed to get that straight and more importantly, parallel grain" [D8]. Every dedicated dowel-making tool in this entry is downstream of this method, and — the deepest point in this entry — the tools mechanized the *shape* while quietly making it possible to abandon the *grain discipline* that gave the shape its strength. - **Dowels are documented in ancient Egyptian joinery.** Ancient Egyptian joinery used dowels structurally: locked mortise-and-tenon joints where a loose tenon is "locked in place by driving a dowel through one or more holes," and butterfly keys locked with paired dowels — with surviving artifacts catalogued by number, including Tutankhamun's shrine GEM No. 21072 (18th Dynasty), whose reused mortises were patched "with another piece of wood secured with dowels," and Old Kingdom bed frames in the Cairo Museum (Inventory No. 480) [D12]. **How those dowels were made is not described in the source — an honest gap at the ancient end: the artifacts are proof of early dowel use only; the shaping tool and production method remain unknown.** The source documents dowelled objects, not a dowel plate, riven preparation or any production sequence, and a dowel-secured repair does not show that dowels alone held an object together; this entry draws no lineage from the ancient objects to any tool below. - **The rounding plane / witchet — the hand-tool ancestor of the rotary form.** A body with a round, usually tapered channel and an angled iron, rotated around the stock "like a large pencil shaver" [D6]. A dated physical example: a witchet marked **T. Turner, Queen St., Sheffield, dated 1846–1858** (Turner active c. 1840–1912), documented in Lee Valley's own tool-history column, which also traces the word "witchet" to its first print appearance in **Henry C. Mercer's *Ancient Carpenters' Tools* (1929)** [D7]. The alternate names are themselves lineage evidence: *trenail rounder (moot)* ties the form to shipbuilding treenails, *thole plane* to thole pins — the tool followed the peg into every trade that needed one [D7]. - **The Stanley No. 77 dowel and rod turning machine** — the industrialized middle node: a hand-cranked benchtop machine with interchangeable cutter heads, dealer-dated to **1911–1969** [D13, MEDIUM — dealer listing; Stanley's own support page acknowledges the machine but its text was not fetched this pass]. A Lost Art Press commenter still owns one, noting "the cutters for them are so expensive its not economical" [D11] — a living tool priced out of use, not a dead one. **Contemporaries** - **The lathe** — the general-purpose contemporary. Makes any cylinder, plus the tapers and swells no dowel maker can; slower per piece, length-limited by the bed, and demands turning skill. The [[Trapping Plane]] and [[Stail Engine]] historically worked *with* lathes [D6]. *(Stail engine now built as its own entry, Run 642 — an adjustable/threaded-arm witchet variant; the shared "Fred Lambert" inventor lead investigated but left unresolved.)* *(Trapping plane now built as its own entry, Run 641 — a named-inventor, treadle/powered-lathe chairmaking-course accessory, distinct from the pole-lathe bodging tradition; see that entry.)* - **The [[Dowel & Tenon Cutter (Hollow Auger)|dowel & tenon cutter and tapered tenon cutter]]** (Veritas and others; Ray Iles, Ashem Crafts rotary planes in the chairmaking world) — siblings that cut a tenon on the *end* of a stick rather than a full-length dowel; the tapered form has useful self-seating geometry and is favoured by some chairmakers; one enthusiast blog says "a tapered tenon becomes tighter over time if force is applied against it, and has little chance of pushing through mortise over time like a round tenon will" [D6, MEDIUM — one maker's claim, not joint doctrine; straight cylindrical tenons are used successfully in chair and furniture work when properly sized and fitted]. *(Now built as its own entry, Run 533 — which found a named working chairmaker directly disputing this tightening claim; see that entry's History section.)* - **The Veritas Dowel Former** — a modern hybrid of the two live forms: a pound-through tool like a plate, but with a true shearing blade [D1-adjacent; product line confirmed, instruction sheet not fetched this pass]. **Descendants / the live patent chain** - **US 6,263,929 B1, "Dowel maker" — fetched and read directly.** Inventors **Abdul Aziz, Leonard G. Lee, Lloyd Sevack**, assignee **Lee Valley Tools Ltd**, filed August 26, 1999, granted July 24, 2001 [D3]. Its own prior-art citations run the family back through the machine age: Hancharik's "Dowel making machine" (US 2,529,018, 1950), Brown's "Dowel cutter and sizer" (US 4,553,575, 1985), Hsieh's "Wood bar rounding machine" (US 4,770,215, 1988) [D3]. - **Later patent records cite US 6,263,929 B1**, including a Wittner GmbH peg-cutter application (US20070214938A1, 2007), a portable dowel-making tool granted 2022 and a tenon-tapering-plane iron-setting key published 2024 [D3]. A forward citation shows technical relevance, not copying, direct descent or identity of mechanism; comparing the later patents' claims and mechanisms would be required before calling any one a descendant, and the later patents were not read here. The lineage is not claimed closed. (Google's record lists the patent as expired; legal-status labels are informational, not a legal opinion.) --- ## history **The Veritas patent, read for what it is: the applicants' own statement of the problem.** The patent record for US 6,263,929 B1, read directly in the earlier pass, has a background section that states what the inventors saw as the problem. In the applicants' words, commercial dowel-making machines were "complex and expensive," and commercial dowels "vary wildly from their intended nominal diameters, are often oval rather than round in cross-section, and are available in a limited number of species." That is the inventors' background and invention rationale, not independent market testing of commercial dowel: it is primary evidence of what they claimed and disclosed, and broader claims about commercial dowel would need measurement data, standards or tolerances, or a multi-brand sample. Single-blade tools, the inventors say, embed a trade-off: "there is always a trade-off between the 'quantity' of a cut... and the quality of the surface produced" — which the invention answers with two blades, a roughing cut and a fine shearing cut, each on its own micro-adjust [D3]. Leonard G. Lee is named as an inventor, between Abdul Aziz and Lloyd Sevack; his company role is not something the patent proves, and would need Lee Valley's own corporate history to cite [D3]. **The tool's marketing makes the same case in plainer language, as manufacturer marketing rather than independent measurement.** Lee Valley's own product page: "making accurately sized dowels has been beyond the ability of the small shop," and commercial dowel is "imprecise at best and wildly inaccurate at worst" [D1]. A Sawmill Creek user, Dennis Droege, reported that he wanted 3/8" dowels and "noticed commercial dowels vary in size" [D5]. Practitioner reports show that undersized or oval dowels occur; they do not show that commercial dowel is generally defective. **The plate's history is older in form and murkier in record.** The Lie-Nielsen plate — A2 tool steel "hardened to 60 RC," cryogenically treated, seven holes 1/8" to 5/8", each with "a 6° clearance taper on the underside" and a straight land "for the first .045″" so it can be resharpened without changing size [D2] — is a precision modern rendering of an old shop object. But this pass could not establish *when the graduated hardened dowel plate became a woodworking product* (as opposed to a blacksmith's drawplate for metal, a nail header, or a shop-made hole-in-hard-steel expedient). **Named plainly as unresolved**: the plate's form is ancient-adjacent; its documented history as a commercial woodworking tool is thin in the record this pass reached, and Salaman — the canon source most likely to settle the witchet/rounder/plate terminology and dates — was not directly consulted (see *confidence*). **The drawbore tradition is the reason the family stayed alive.** Joseph Moxon's *Mechanick Exercises* (17th century) prescribed offsetting the peg holes "by the width of a shilling" — about 1/16" — so driving the peg pulls the joint tight; Christopher Schwarz's revival argument is blunt: "Drawboring is one of the simple reasons that so much antique furniture survives today, some of it as sound as the day it was made" [D10, Moxon relayed via Schwarz]. A drawbored peg is bent inside the joint, permanently, under load — which is why riven, demonstrably straight-grained stock is strongly preferred for highly bent or loaded pegs, and why any dowel (riven, sawn, turned or purchased) should be inspected or specified for sufficiently straight, continuous grain. The whole family's revival case compresses into one sentence: **the peg is a load-carrying and alignment-critical part of a drawbored joint; its grain continuity, species, moisture, diameter and fit must suit the joint. It does not substitute for sound mortise-and-tenon geometry or structural design.** **The witchet's fade and the rounder's survival.** The adjustable rounding plane (stail engine) made "broom handles, snooker cues, fishing rods" [D6]; the trades that consumed handles by the thousand industrialized, and the tool faded with them. But the rounder survived in exactly one living tradition — chairmaking — where Ashem Crafts still manufactures rotary planes in England today [D14], and where the modern rotary dowel maker has re-entered: the Veritas is used on James Mursell's Windsor chairmaking courses [D11, commenter attribution — see *human side*]. --- ## ✦ human side **Michael Peet's disappointment — the family's defining expectation gap, in one thread read in full.** Sawmill Creek, "Dowel Plate Woes": "I ripped open the package excitedly and ran down to the shop to make some dowels... banged it through the 1/4" hole. Hmm, not too straight. Or smooth. Or round." He'd tried cedar and oak, several hammers, and asked honestly: "Am I supposed to be using greener wood? Should I have lapped the plate first?" [D4]. The thread's named answers form a complete folk curriculum: **Harry Strasil** on strike mechanics — "hit a nail on one side and it bends away from the blow, same with making dowels with a plate, you have to hit it fair and square"; **Charles Murray**, on his view of the mechanism (the maker describes a cutting surface; his is a practitioner's characterization) — "a dowel plate will work much easier with hardwoods (remember it's a scraping action not cutting)" and, as his own technique, stepping down through the holes, "this takes a little off at a time"; **David Gendron** and **Tony Zaffuto** on riven, drawknifed, octagoned prep; **Chuck Nickerson** on the honest fix — "lowering my expectations... I found a work surface at belt height. That solved my problem"; **Russ Hauser** on mounting the plate at an angle in an Emmert vise so both hands are free; **Don Dorn** on the cleverest trick in the thread — buy a commercial dowel of the species you need and run *it* through the plate: "It goes in easily but not perfect, but comes out that way" [D4]. **scrollsawguy's $400 ragged stick — the rotary form's own disaster account.** SawdustZone: after buying the dowel maker to make decorative canes, "ended up with a ragged tornout lookin stick that there is NO way to call it a dowel," despite following the instructions repeatedly across species [D9]. The named fixes came from **Uncle Cracker**, who runs 1"+ martial-arts stick blanks and laminates through his: "trim the stock down on 8 sides, instead of 4" to unload the roughing blade, drive it with a Milwaukee 1/2" corded drill (his preference, not a Veritas requirement), expect more tearout in softer woods; and **Slik Geek**'s diagnostic triage — feed too fast? blade sharp? stock actually 1/8" oversize? [D9]. **Christopher Schwarz, deciding when the tool earns its place.** Asked whether the Veritas suits chairmaking: yes, for production — "I could select the straightest, clearest stock to make spindles that were super strong... you need 50 perfect dowels with dead-straight grain, the tool is a nice thing to have." And no, for the occasional chair — "I'd make the sticks the old-fashioned way with a spokeshave or block plane. Saw them out straight. Make them octagons. Plane them round" [D11]. A commenter adds that this "is the same dowel maker **James Mursell** uses on his Windsor chair making courses. Works great but a corded drill is advisable though!" [D11 — commenter attribution, not Mursell's own words; held at MEDIUM]. Another commenter, on the tool's notorious documentation: "I'm glad it was not just me who struggled to find useful instructions for this thing" [D11]. **Warren Mickley on species matching — a practitioner's rule of thumb, kept as an example, not a specification.** On drawbore pin species: "A white oak pin will just blow right through a white pine joint" — a practitioner's account of historic practice, not a design value; the right peg species depends on peg diameter, species and density, member material, joint geometry, moisture, load and desired failure mode [D13-thread; Timber Framers Guild curriculum, per the review, relayed]. **Derek Cohen** supplies the flex principle: "The pin needs to have some flex... Rigid, short-grained wood is likely to crack"; **Tom M King** carries the historical failure record — aggressive drawboring in dry wood, pegs blowing out tenons; **Jim Koepke** his numbers (offset to 1/32", pin spacing minimums — one maker's scale-specific figures, not to be transferred between scales); and **Charles Guest** the dissent this entry preserves rather than smooths over: drawboring suits massive timber frames and is "vastly overrated" for modern furniture [D13-thread]. **Steve Hubbard** rounds out the revival motive from the Veritas side: he bought his to make dowels from padauk and hard maple, "species that are difficult to obtain in commercially available dowels" [D5]. --- ## ✦ proof of existence - **US Patent 6,263,929 B1, "Dowel maker" — fetched and read directly, in full, this run.** Inventors Abdul Aziz, Leonard G. Lee, Lloyd Sevack; assignee Lee Valley Tools Ltd; filed 1999-08-26, granted 2001-07-24; with a prior-art chain to 1950 (Hancharik US 2,529,018) and forward citations to 2024, including a Wittner peg-cutter application (2007) that cites it (citation is not descent). **HIGH for the exact content of the patent as read;** its background is the applicants' account, not independent proof [D3]. - **Ancient dowelled artifacts, catalogued by museum number — via a directly-read scholarly chapter.** Tutankhamun shrine GEM No. 21072 with dowel-secured patches (18th Dynasty); Old Kingdom bed frames, Cairo Museum Inventory No. 480; locked mortise-and-tenon and dowel-locked butterfly joints described; author Medhat Abdallah Abd Elhamid (Saqqara storerooms conservation). **Proof of early dowel use only; the shaping tool and production method remain unknown.** MEDIUM-HIGH for the objects as the chapter describes them; the making is an honest gap, and nothing here links these objects to a dowel plate or any modern tool [D12]. - **A dated physical witchet**: T. Turner, Queen St., Sheffield, 1846–1858 (maker active c. 1840–1912), documented with the word's first print appearance (Mercer, *Ancient Carpenters' Tools*, 1929) in Lee Valley's own tool-history column — directly read. **MEDIUM-HIGH** (columnist-documented example, not a museum accession) [D7]. - **Moxon's 17th-century drawbore offset** ("the width of a shilling") — relayed via Schwarz's article, not read in Moxon directly. **MEDIUM-HIGH** [D10]. - **Stanley No. 77 dowel and rod turning machine, 1911–1969** — dealer-dated (Tooltique listing; Stanley's own support page acknowledges the machine; a surviving user in the LAP comments). **MEDIUM** — no Stanley-primary date document read this pass [D13]. - **Living manufacture, both live forms**: Lie-Nielsen Standard Dowel Plate (1-DPS) and Veritas Dowel Maker (05J4501/05J4520) — both manufacturers' own product literature fetched and read directly. **HIGH** [D1][D2]. - **Museum accession record for a dowel plate specifically**: none located this pass — named as a gap, not an absence. --- ## references [D1] Lee Valley Tools, Veritas Dowel Maker product page (05J4501/05J4520) — https://www.leevalley.com/en-us/shop/tools/hand-tools/dowel-and-tenon-cutters/42331-veritas-dowel-maker | kind: manufacturer-primary (HIGH for product specifications as stated; MEDIUM for marketing claims such as "perfectly sized" and "any species" — fetched and read directly in the earlier pass and again in the fix pass) [D2] Lie-Nielsen Toolworks, Standard Dowel Plate (1-DPS) product page — https://www.lie-nielsen.com/products/dowel-plate-standard | kind: manufacturer-primary (HIGH for product specifications as stated — read directly in the earlier pass; in the fix pass the metric-plate page, lie-nielsen.com/products/dowel-plates-dowel-plate-metric-, was read: its specification says 1/4" thick and its instructions say 5/16", unresolved) [D3] Abdul Aziz, Leonard G. Lee, Lloyd Sevack, "Dowel maker," US Patent 6,263,929 B1 (filed 1999-08-26, granted 2001-07-24), assignee Lee Valley Tools Ltd, including background section, prior-art citations (US 2,529,018 1950; US 4,553,575 1985; US 4,768,903 1988; US 4,770,215 1988; US 5,127,452 1992) and full Cited-By table (11 patents, 2007–2024, incl. Wittner GmbH US20070214938A1) — https://patents.google.com/patent/US6263929B1/en | kind: patent-primary (HIGH for the exact content of the patent as read; the background section is the applicants' account, not independent proof) [D4] "Dowel Plate Woes," Sawmill Creek Woodworking Community (archive), started by Michael Peet, with named replies from Harry Strasil, Chuck Nickerson, Harlan Barnhart, Charles Murray, David Gendron, Tony Zaffuto, Russ Hauser, Don Dorn, Eric Brown, Allan Brown — https://sawmillcreek.org/archive/index.php/t-120302.html | kind: forum (MEDIUM — attributed practitioner evidence, read in full in the earlier pass; named first-person experience, not verification of tool limits) [D5] "Veritas dowel maker," Sawmill Creek Woodworking Community, started by Dennis Droege, with named replies from Stephen Rosenthal, Todd Trebuna, Steve Hubbard, John Gornall — https://sawmillcreek.org/threads/veritas-dowel-maker.287971/ | kind: forum (MEDIUM — attributed practitioner evidence, read in the earlier pass) [D6] "Rounding Planes," Vintage Tool Patch blog — https://blog.vintagetoolpatch.com/2022/04/16/rounding-planes.html | kind: enthusiast blog (MEDIUM — competent and specific, but cites no historical dictionaries and no dated makers; fetched and read directly this run) [D7] "What Is It? — the Witchet," Lee Valley Tools tool-history column — https://www.leevalley.com/en-us/discover/what-is-it-articles/whatisit-witch | kind: manufacturer tool-history column, primary-adjacent (MEDIUM-HIGH — fetched and read directly this run; carries the T. Turner 1846–58 dated example, the full alternate-name list, and the Mercer 1929 etymology) [D8] Richard Maguire, "How To Make Strong Wooden Pegs Without A Dowel Plate," The English Woodworker — https://www.theenglishwoodworker.com/how-to-make-wooden-pegs/ | kind: named-practitioner blog (MEDIUM — attributed practitioner evidence, read in the earlier pass) [D9] "lee valley dowel maker,,help!!!," The SawdustZone, started by scrollsawguy, with named replies from Slik Geek and Uncle Cracker — https://www.sawdustzone.org/forum/discussions/tool-talk/31476-lee-valley-dowel-maker-help | kind: forum (MEDIUM — first-person failure testimony, attributed practitioner evidence, read in the earlier pass) [D10] Christopher Schwarz, "Drawboring Resurrected," Popular Woodworking (orig. Woodworking Magazine, Autumn 2005), citing Joseph Moxon, Mechanick Exercises (17th c.) — https://www.popularwoodworking.com/techniques/drawboring-resurrected/ | kind: named-author trade press citing a period primary (MEDIUM — read in the earlier pass; Moxon relayed, not read in the original) [D11] Christopher Schwarz, "Veritas Dowel Maker for Chairmaking?," Lost Art Press blog, 2019-09-01, incl. reader comments (James Mursell attribution; Stanley No. 77 owner; instructions complaint) — https://blog.lostartpress.com/2019/09/01/veritas-dowel-maker-for-chairmaking/ | kind: named-author craft press + comments (MEDIUM for Schwarz's own words as practitioner opinion; lower for commenter attributions — read in the earlier pass) [D12] Medhat Abdallah Abd Elhamid, "Woodworking Techniques in Ancient Egypt," IntechOpen — https://www.intechopen.com/chapters/1201517 | kind: scholarly chapter with museum catalog numbers (MEDIUM-HIGH for the object descriptions; it supports dowelled joinery in specific objects and fails to support any dowel-making tool, riving method or lineage to a modern tool) [D13] Stanley No. 77 dating: Tooltique dealer listing ("1911–1969") — https://www.tooltique.co.uk/shop/rare-vintage-stanley-usa-no-77-benchtop-dowel-rod-turning-machine-1911-1969-good-condition/ ; Stanley Tools support page acknowledging the No. 77 — https://support.stanleytools.com/hc/en-us/articles/360012685998 | kind: dealer listing + manufacturer support page (MEDIUM — neither fetched in full; dates not manufacturer-primary-verified) [D13-thread] "Draw Bore Question - Pin Species," Sawmill Creek Woodworking Community (archive), started by Joe A Faulkner (2021-07-09), with named replies from Derek Cohen, Prashun Patel, Warren Mickley, Rob Luter, Tom M King, Charles Guest, Jim Koepke — https://sawmillcreek.org/archive/index.php/t-292513.html | kind: forum (MEDIUM — attributed practitioner evidence, read in the earlier pass) [D14] Ashem Crafts, rotary planes (living UK manufacturer of rounders for chairmaking) — https://www.ashemcrafts.com/ | kind: manufacturer (MEDIUM — located and confirmed live via search this pass; site not read in depth) [D15] Derek Cohen, "Drawboring," inthewoodshop.com — https://www.inthewoodshop.com/Furniture/Drawboring.html | kind: named-practitioner site (MEDIUM — read in the earlier pass; documents the riven-split-chamfer-plate workflow and the LN plate's hole geometry in use) [D16] Veritas Dowel Maker official manual — https://assets.veritastools.ca/Original/10091/42331-veritas-dowel-maker-c-01-e.pdf | kind: manufacturer-primary (**not read in the fix pass:** the fetch returned an error. The reviewer's 2026-07-29 review reports reading it in full; its safety controls, blank-size rule, drill and adapter wording, speed and feed guidance, support intervals and length limit are carried here as `[manual, relayed]`.) [D17] CCOHS, "Hand Tools — Struck Tools" — https://www.ccohs.ca/oshanswers/safety_haz/hand_tools/struck.html | kind: government safety authority (HIGH for the general struck-tool guidance, read in the fix pass: eye and face protection, screens for bystanders, correct-tool selection, inspect heads for mushrooming or chips; it does not mention dowel plates or prescribe a mallet type) [D18] Timber Framers Guild, training curriculum outlines — https://www.tfguild.org/downloads/CompleteCurriculumOutlines.pdf | kind: trade-guild curriculum (relayed from the review, not read: it treats peg species, grading, form, moisture, sizing, grain orientation and drawboring as engineering and practice variables) [D19] US Forest Products Laboratory, Wood Handbook, Chapter 8: Fastenings — https://research.fs.usda.gov/treesearch/62253 | kind: government research (relayed from the review, not read: fastening design must be adapted to wood strength and dimensional change; no universal drawbore-peg formula) --- ## The reviewer's corrections (15) - `BLOCKER` — 1. Replace the incomplete rotary-machine safety section with the official manual’s controls — Exact claim: The Veritas manual “could not be fetched,” so the entry infers torque and whipping hazards primarily from comments and says no manufacturer warning could be quoted. - `BLOCKER` — 2. Add the manufacturer’s diameter-specific support intervals and length limit — Exact claim: Long blanks can whip, but the entry gives no manufacturer support schedule and markets dowels “to 4 ft (or longer)” as general capability. - `BLOCKER` — 3. Correct the square-stock sizing rule — Exact claim: “Rip blanks square, 1/8-inch oversize,” repeated as the universal rotary-maker input. - `SHOULD-FIX` — 4. Stop prescribing a corded 1/2-inch drill as though it were the manufacturer’s requirement — Exact claim: The rotary form pairs with and should be driven by “a corded 1/2-inch drill.” - `BLOCKER` — 5. Correct the dowel plate’s “step-down ladder” and scraping-only mechanism — Exact claim: “The working sequence is a step-down ladder, not a single die”; “the plate scrapes rather than cuts”; and users should “step down through the graduated holes.” - `SHOULD-FIX` — 6. Remove the universal claim that a dowel plate makes only rough structural pegs — Exact claim: “The plate makes structural pegs; it does not make showroom dowels,” and it should never be used for long finish dowels. - `SHOULD-FIX` — 7. Correct plate specifications and durability language — Exact claim: The plate is 1/4 inch thick and “hard enough to shear hardwood indefinitely.” - `BLOCKER` — 8. Stop treating patent background as independent “under oath” proof — Exact claim: The patent “names the failures the form exists to fix, in writing, under oath of novelty,” proving commercial dowels “vary wildly,” are oval, and are species-limited. - `SHOULD-FIX` — 9. Do not infer direct descendants from patent citations — Exact claim: The patent’s “forward citations show the family still branching,” including Wittner’s peg cutter as “the same mechanism migrated to violin tuning pegs.” - `BLOCKER` — 10. Remove the ancient-Egypt tool-lineage contradiction — Exact claim: The entry first says the ancient artifacts’ dowel-production method is unknown, then concludes that driving a riven peg through a modern dowel plate is “exactly what held Tutankhamun’s shrines together.” - `BLOCKER` — 11. Replace “the peg IS the structure” with accurate joint mechanics — Exact claim: “In a drawbored joint, the peg is not a fastener holding the structure—the peg IS the structure.” - `BLOCKER` — 12. Remove universal hardwood-only, species-pairing, and drawbore-offset rules — Exact claim: “Hardwoods only for structural pegs”; softwood structural pegs are a job the tool “should never do”; a white-oak pin will “blow right through” white pine; and working offsets run from 1/32 to 3/32 inch. - `BLOCKER` — 13. Remove the universal claim that straight tenons work loose or push through — Exact claim: A tapered tenon is required because a straight round tenon “will” push through or work loose, while the dowel maker is categorically wrong for chair tenons. - `SHOULD-FIX` — 14. Bound the riven-versus-commercial grain claims — Exact claim: A machine/commercial dowel “sawn from the board regardless of grain direction” carries hidden runout and reintroduces a failure that the riven ancestor solved. - `BLOCKER` — 15. Repair the schema placeholders and confidence/source grading — Exact claim: `uid: haft-dowel-maker-{{RUN_NO}}`, `haft_run_id: Run {{RUN_NO}}`, the title note repeats `{{RUN_NO}}`, and several forum threads are graded HIGH for broad technical claims.